AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation

Tal Reiss, Niv Cohen, Liron Bergman, Yedid Hoshen

CVPR 2021 · Label-efficient and robust learning › OOD and anomaly detection · approach: Unified multi-class anomaly detection · 285 citations (OpenAlex)

Check your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Adapt ImageNet-pretrained features with compactness training and elastic regularizationPANDA on CIFAR10 · CIFAR10ROC AUC (%): 92.5 -> 96.2Improved
Adapt ImageNet-pretrained features with compactness training and elastic regularizationPANDA on CatsVsDogs · CatsVsDogsROC AUC (%): 96.0 -> 97.3Improved
Add outlier exposure using 80M Tiny ImagesPANDA on CIFAR10 · CIFAR10ROC AUC (%): 96.2 -> 98.9Improved
Add outlier exposure using 80M Tiny ImagesPANDA on Fashion MNIST · Fashion MNISTROC AUC (%): 95.6 -> 91.8Hurt
Use pretrained features to initialize RotNet (ablation)MHRot on CIFAR10 · CIFAR10ROC AUC (%): 90.1 -> 86.6Hurt
Use PANDA-OE rather than MHRot with outlier exposureCIFAR10 outlier exposure · CIFAR10ROC AUC (%): 95.6 -> 98.9Improved
Use SPADE pretrained multiscale pixel features and nearest-neighbor scoringMVTec anomaly segmentation · MVTecpixel-level ROC AUC (%): 89 -> 96.2Improved
Use sample-wise rather than constant-duration early stopping (ablation)PANDA on CatsVsDogs · CatsVsDogsROC AUC (%): 91.9 -> 95.7Improved
Use elastic weight consolidation rather than constant-duration early stopping (ablation)PANDA on CatsVsDogs · CatsVsDogsROC AUC (%): 91.9 -> 97.3Improved
Use elastic weight consolidation rather than constant-duration early stopping (ablation)PANDA on CIFAR100 · CIFAR100ROC AUC (%): 94.8 -> 94.1Hurt
Finetune all four ResNet blocks instead of blocks 3 and 4 (ablation)PANDA-EWC on CIFAR10 · CIFAR10ROC AUC (%): 96.2 -> 94.9Hurt
Use only ResNet layer 0 rather than layers 0, 1, and 2 (ablation)SPADE on MVTec · MVTecPRO (%): 97.3 -> 90.6Hurt

Later papers that compared with this method

6 later accepted papers report PANDA as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy